The MPLS WG Archive

[Ping Pan <pingpan@cs.columbia.edu>]

Randy Bush wrote:
> 
> hi ping,
> 
> > From various studies (Vern Paxson...), it was shown that end-to-end
> > jitter can be quite high.
>

Hello, Randy,
 
> i think it is actually less high inter-backbone than seems to be generally
> thought.  e.g. i will append (humble apologies for the non-ascii, but it is
> a picture) a ippm-style ripe traffic measurement between verio denver and
> ripe in amsterdam.  there are outliers.
>

Thanks for the info. Is it the data between two backbones over one NAP?
Is it generally the same for all inter-provider traffic?
 
> > Within core networks, packet delay and loss may be low. But after going
> > through multiple domains, NAP's and POP's
> 
> i do not mean to be pedantic, but i am not aware of how, from a policy
> routing perspective, traffic would go through multiple naps.  i push the
> point because the suckyness (sp?) of some of the naps is one of the worst
> sources of interprovider problems.
> 
> i am not sure what 'domains' are.  if you mean autonomous systems, then
> this devolves to naps, inter-provider private circuits, and intra-provider
> pops.
>

I meant AS's. I first saw the data in http://moat.nlanr.net/ASPL/. I
chatted with one of my colleagues in Bell Labs (he had gone already) a
couple of months ago. He was working on inter-domain network topology
stuff, and collected bunch of more up to date traces. He told me that
the average AS length in his data was 5-6.

My point was that even though major transit ISP's (Verio and UUnet) can
control link congestion and transit delay within their networks, can all
your customers (regional ISP's) provide the same level of service to end
users? If no, what can we do about it? One solution may be to create a
MPLS tunnel with bandwidth guarantees from one regional ISP over the
backbones to another regional ISP...

Thanks!

--
Ping Pan         http://www.cs.columbia.edu/~pingpan